Frequency Domain Unlocks New Perspectives for Abdominal Medical Image Segmentation

2025-10-13

AI Summary

To address the challenges of low tumor-to-normal-tissue contrast and ambiguous boundaries in abdominal medical images—which degrade segmentation accuracy—this paper proposes a foreground-aware frequency-domain segmentation framework. The method introduces a foreground-aware module to enhance discriminability of target regions across the full field of view; incorporates wavelet transform for feature-level frequency-domain enhancement, emphasizing high-frequency details; and designs an edge-constrained loss function to ensure geometric continuity of segmentation boundaries. By synergistically integrating spatial and frequency-domain information, the model significantly improves fine-grained structural segmentation under complex anatomical backgrounds. Extensive experiments on multiple public abdominal medical image datasets demonstrate state-of-the-art performance in key metrics—including Dice score and 95th-percentile Hausdorff Distance (HD95)—particularly excelling in low-contrast and small-lesion segmentation tasks.

Technology Category

• Computer Vision: Segmentation
• Knowledge Representation and Reasoning: Diagnosis and Abductive Reasoning
• Machine Learning: Multi-instance/Multi-view Learning

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods

Abstract

Accurate segmentation of tumors and adjacent normal tissues in medical images is essential for surgical planning and tumor staging. Although foundation models generally perform well in segmentation tasks, they often struggle to focus on foreground areas in complex, low-contrast backgrounds, where some malignant tumors closely resemble normal organs, complicating contextual differentiation. To address these challenges, we propose the Foreground-Aware Spectrum Segmentation (FASS) framework. First, we introduce a foreground-aware module to amplify the distinction between background and the entire volume space, allowing the model to concentrate more effectively on target areas. Next, a feature-level frequency enhancement module, based on wavelet transform, extracts discriminative high-frequency features to enhance boundary recognition and detail perception. Eventually, we introduce an edge constraint module to preserve geometric continuity in segmentation boundaries. Extensive experiments on multiple medical datasets demonstrate superior performance across all metrics, validating the effectiveness of our framework, particularly in robustness under complex conditions and fine structure recognition. Our framework significantly enhances segmentation of low-contrast images, paving the way for applications in more diverse and complex medical imaging scenarios.

Problem

• Segmenting tumors in low-contrast abdominal medical images
• Differentiating malignant tumors resembling normal organs
• Enhancing boundary recognition in complex medical backgrounds

Innovation

• Foreground-aware module amplifies background distinction for target focus
• Wavelet transform extracts high-frequency features for boundary enhancement
• Edge constraint module preserves geometric continuity in segmentation